Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images

The estimation of Above-Ground Biomass (AGB) in Amorphophallus konjac (Konjac) is essential for field management and yield prediction. While previous research has demonstrated the efficacy of Unmanned Aerial Vehicle (UAV) RGB imagery in estimating AGB for monoculture crops, the applicability of thes...

Full description

Saved in:
Bibliographic Details
Main Authors: Ziyi Yang, Hongjuan Qi, Kunrong Hu, Weili Kou, Weiheng Xu, Huan Wang, Ning Lu
Format: Article
Language:English
Published: MDPI AG 2025-03-01
Series:Drones
Subjects:
Online Access:https://www.mdpi.com/2504-446X/9/3/220
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850204042018422784
author Ziyi Yang
Hongjuan Qi
Kunrong Hu
Weili Kou
Weiheng Xu
Huan Wang
Ning Lu
author_facet Ziyi Yang
Hongjuan Qi
Kunrong Hu
Weili Kou
Weiheng Xu
Huan Wang
Ning Lu
author_sort Ziyi Yang
collection DOAJ
description The estimation of Above-Ground Biomass (AGB) in Amorphophallus konjac (Konjac) is essential for field management and yield prediction. While previous research has demonstrated the efficacy of Unmanned Aerial Vehicle (UAV) RGB imagery in estimating AGB for monoculture crops, the applicability of these methods to AGB estimation in Konjac remains uncertain due to its distinct morphological traits and prevalent intercropping practices with maize. Additionally, the Vegetation Indices (VIs) and Texture Features (TFs) obtained from UAV-based RGB imagery exhibit significant redundancy, raising concerns about whether the selected optimal variables can maintain estimation accuracy. Therefore, this study assessed the effectiveness of Variable Selection Using Random Forests (VSURF) and Principal Component Analysis (PCA) in variable selection and compared the performance of Stepwise Multiple Linear Regression (SMLR) with four Machine Learning (ML) regression techniques: Random Forest Regression (RFR), Extreme Gradient Boosting Regression (XGBR), Partial Least Squares Regression (PLSR), and Support Vector Regression (SVR), as well as Deep Learning (DL), in estimating the AGB of Konjac based on the selected features. The results indicate that the integration (PCA_(PCA_VIs+PCA_TFs)) of PCA-based VIs and PCA-based TFs using PCA achieved the best prediction accuracy (R<sup>2</sup> = 0.96, RMSE = 0.08 t/hm<sup>2</sup>, MAE = 0.06 t/hm<sup>2</sup>) with SVR. In contrast, the DL model derived from AlexNet, combined with RGB imagery, yielded moderate predictive accuracy (R<sup>2</sup> = 0.72, RMSE = 0.21 t/hm<sup>2</sup>, MAE = 0.17 t/hm<sup>2</sup>) compared with the optimal ML model. Our findings suggest that ML regression techniques, combined with appropriate variable-selected approaches, outperformed DL techniques in estimating the AGB of Konjac. This study not only provides new insights into AGB estimation in Konjac but also offers valuable guidance for estimating AGB in other crops, thereby advancing the application of UAV technology in crop biomass estimation.
format Article
id doaj-art-04c30d22e4c54d51aee4218ceac7c7b1
institution OA Journals
issn 2504-446X
language English
publishDate 2025-03-01
publisher MDPI AG
record_format Article
series Drones
spelling doaj-art-04c30d22e4c54d51aee4218ceac7c7b12025-08-20T02:11:22ZengMDPI AGDrones2504-446X2025-03-019322010.3390/drones9030220Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB ImagesZiyi Yang0Hongjuan Qi1Kunrong Hu2Weili Kou3Weiheng Xu4Huan Wang5Ning Lu6College of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaAgricultural and Rural Development Service Center of Housuo Town, Qujing 655501, ChinaCollege of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaCollege of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaCollege of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaCollege of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaCollege of Big Data and Intelligent Engineering, Southwest Forestry University, Kunming 650223, ChinaThe estimation of Above-Ground Biomass (AGB) in Amorphophallus konjac (Konjac) is essential for field management and yield prediction. While previous research has demonstrated the efficacy of Unmanned Aerial Vehicle (UAV) RGB imagery in estimating AGB for monoculture crops, the applicability of these methods to AGB estimation in Konjac remains uncertain due to its distinct morphological traits and prevalent intercropping practices with maize. Additionally, the Vegetation Indices (VIs) and Texture Features (TFs) obtained from UAV-based RGB imagery exhibit significant redundancy, raising concerns about whether the selected optimal variables can maintain estimation accuracy. Therefore, this study assessed the effectiveness of Variable Selection Using Random Forests (VSURF) and Principal Component Analysis (PCA) in variable selection and compared the performance of Stepwise Multiple Linear Regression (SMLR) with four Machine Learning (ML) regression techniques: Random Forest Regression (RFR), Extreme Gradient Boosting Regression (XGBR), Partial Least Squares Regression (PLSR), and Support Vector Regression (SVR), as well as Deep Learning (DL), in estimating the AGB of Konjac based on the selected features. The results indicate that the integration (PCA_(PCA_VIs+PCA_TFs)) of PCA-based VIs and PCA-based TFs using PCA achieved the best prediction accuracy (R<sup>2</sup> = 0.96, RMSE = 0.08 t/hm<sup>2</sup>, MAE = 0.06 t/hm<sup>2</sup>) with SVR. In contrast, the DL model derived from AlexNet, combined with RGB imagery, yielded moderate predictive accuracy (R<sup>2</sup> = 0.72, RMSE = 0.21 t/hm<sup>2</sup>, MAE = 0.17 t/hm<sup>2</sup>) compared with the optimal ML model. Our findings suggest that ML regression techniques, combined with appropriate variable-selected approaches, outperformed DL techniques in estimating the AGB of Konjac. This study not only provides new insights into AGB estimation in Konjac but also offers valuable guidance for estimating AGB in other crops, thereby advancing the application of UAV technology in crop biomass estimation.https://www.mdpi.com/2504-446X/9/3/220konjacabove-ground biomassregression techniquesUAV-based RGBdeep learning
spellingShingle Ziyi Yang
Hongjuan Qi
Kunrong Hu
Weili Kou
Weiheng Xu
Huan Wang
Ning Lu
Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
Drones
konjac
above-ground biomass
regression techniques
UAV-based RGB
deep learning
title Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
title_full Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
title_fullStr Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
title_full_unstemmed Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
title_short Estimation of Amorphophallus Konjac Above-Ground Biomass by Integrating Spectral and Texture Information from Unmanned Aerial Vehicle-Based RGB Images
title_sort estimation of amorphophallus konjac above ground biomass by integrating spectral and texture information from unmanned aerial vehicle based rgb images
topic konjac
above-ground biomass
regression techniques
UAV-based RGB
deep learning
url https://www.mdpi.com/2504-446X/9/3/220
work_keys_str_mv AT ziyiyang estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT hongjuanqi estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT kunronghu estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT weilikou estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT weihengxu estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT huanwang estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages
AT ninglu estimationofamorphophalluskonjacabovegroundbiomassbyintegratingspectralandtextureinformationfromunmannedaerialvehiclebasedrgbimages